by Mike Masnick

News Reports May Be A Little Quick To Clear Sarasota E-Voting Machines

from the not-quite dept

The Associated Press put out an article late Friday claiming that a study found the e-voting machines used in Florida's Sarasota County the machines had no problems -- despite a large number of missing votes. At least that's what you'd get from reading the article, with a headline that blares: "Audit: Fla. Voting Machines Didn't Err." Of course, that's not exactly what the study found. First of all, the panel of researchers did not study the e-voting machines at all -- but just the source code of the software. There could be plenty of other reasons why the voting machines had problems that couldn't be uncovered just by looking at the source code of the software. And, in fact, the actual report is hardly as forgiving as the AP report makes out. Ed Felten points out that the report actually highlights all kinds of security problems with the software, including plenty of places where a virus could exploit a buffer overflow. It also discovered incredibly weak security, such as a master password that would be relatively easy to guess (only had 256 possibilities).

Of course, that doesn't mean that there was anything malicious going on here. As both the report and Felten point out, that hardly seems likely (especially since if you were to do something malicious, you wouldn't undercount votes, but switch them to hide them better). Instead, as Felten notes, it seems likely that the machines simply screwed up. He suggests "systems that are insecure tend to be unreliable as well -- they tend to go wrong on their own even if nobody is attacking them. Code that is laced with buffer overruns, array out-of-bounds errors, integer overflow errors, and the like tends to be flaky. Sporadic undervotes are the kind of behavior you would expect to see from a flaky voting technology." Once again, if anything, this test has done a disservice to those looking to strengthen the election process. As Felten pointed out when he refused to serve on the panel, limiting what could be tested isn't particularly useful -- and leads to things like the Associated Press declaring that the machines have been vindicated when that's not at all true.

Reader Comments (rss)

(Flattened / Threaded)

  1. identicon
    Cixelsid, Feb 26th, 2007 @ 3:28am


    You'd expect software that is integral to the democratic freedom of a country to at least have some sized buffer checks. Seriously, considering the amount of security and testing involved with military software, why aren't the same protocols being adhered to in this case?

    reply to this | link to this | view in thread ]

  2. identicon
    TotallyTrustworthyVotingDude, Feb 26th, 2007 @ 4:53am

    E Voting Machines CAN be trusted

    Electronic voting machines *can* be made totally secure and totally worthy of trust. But only if you let *me* write the software and setup the machines.

    You can totally trust me not to go rigging any elections or anything, because my mum says I'm honest and anyway I have a totally trustworthy name, TotallyTrustworthyVotingDude !

    If there's anything odd about the results, you can totally take my word for it, no need to verify that extra programs haven't been run, or a different dataset has been uploaded, or that the code running on the machine is the one you're certifying, because I'm TotallyTrustworthVotingDude and I say it's stupid voters!

    reply to this | link to this | view in thread ]

  3. identicon
    _Jon, Feb 26th, 2007 @ 5:31am

    So, the most likely scenario is that the people who wrote or tested the software did not run it through multiple iterations of voting. They probably voted a few times and declared "It works!"

    If they had simulated - say - 100+ voters, the system pro'lly would have auto-reset and lost the last voter or two.

    Load Testing - what a concept.

    That reminds me of an auto company that had a problem with their digital odometer when it was new. When the car go to 32767 miles, it would roll over to 0. So much for *actually* driving the car for endurance testing...

    reply to this | link to this | view in thread ]

  4. identicon
    новини, Feb 26th, 2007 @ 6:57am

    digital revolution

    e-voting must be accomplished with open source standarts, hosted on neutral machines with digital and phisical security. Everything else is a big.. LIE. God bless open source.

    reply to this | link to this | view in thread ]

  5. identicon
    Haapi, Feb 26th, 2007 @ 7:11am

    Re: Crazy

    Yeah, I was just wondering why this kind of software isn't written in ADA or some other milspec language.

    Oh, I know, it is the "marketplace at work."

    reply to this | link to this | view in thread ]

  6. identicon
    Anonymous Coward, Feb 26th, 2007 @ 8:28am

    if you think voting machines can be buggy..

    has anyone anywhere found software nearly as capable, OR as compromisable as the humane personality? THAT is what we trust with current voting. Let's keep this in perspective as we responsibly consider our voting future.

    reply to this | link to this | view in thread ]

  7. identicon
    Cixelsid, Feb 26th, 2007 @ 11:35am

    Re: Re: Crazy

    "Yeah, I was just wondering why this kind of software isn't written in ADA or some other milspec language."

    Fuckin A man. You want type safety, fuck, here, you got it. Here's to ADA *clink*

    reply to this | link to this | view in thread ]

Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Use markdown for basic formatting. HTML is no longer supported.
  Save me a cookie
Follow Techdirt
Insider Shop - Show Your Support!

Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Report this ad  |  Hide Techdirt ads
Recent Stories
Report this ad  |  Hide Techdirt ads


Email This

This feature is only available to registered users. Register or sign in to use it.